摘要

The continuous adjoint method for the Reynolds-averaged Navier-Stokes (RANS) equations offers an efficient approach for computing gradients of a functional, particularly in high-dimensional design problems where gradient calculations can otherwise be expensive and difficult to obtain. Existing work has generally been limited to the use of a single, simple functional in the formulation of the adjoint. As a result, the full potential of adjoint-based design has not yet been realized. This work focuses on the derivation and implementation of an adjoint formulation of the RANS equations for generalized outflow-based functionals that allows information from external models to be coupled with the high-fidelity gradient calculation, and expands the range and complexity of objectives that can be used. As an example application this is applied to supersonic combustion ramjets (scramjets), which have the potential to enable more efficient transatmospheric flight, airplane-like operations of launch vehicles, and defense-related applications. Challenges in their design can be addressed by optimization as well as by other gradient-dependent techniques. Optimization studies using the generalized outflow-based functional show that different inlet design results are obtained when incorporating full-engine performance as compared with an inlet-only objective, and illustrate a trade-off between inlet surface heat flux and thrust.

  • 出版日期2017-11